The Ocular Hypertension Treatment Study (OHTS) Supported by the National Eye Institute, National Center on Minority Health and Health Disparities, Research to Prevent Blindness, and Merck Research Laboratories

June, 2002 Ocular Hypertension  Elevated IOP in the absence of clinically detectable optic nerve or visual field changes  A common finding  What to do? –Treat all? –Treat no one? –Treat some? Then who?

June, 2002 Why did we do this study? Don’t we know that treatment prevents open angle glaucoma?

June, 2002 Does Treatment of Ocular Hypertension prevent POAG? InvestigatorProtective Grahamno Norskovno Leveneno David et al. no Chisholmno Schulzer et al. no Heijl et al. no Kamal et al. no InvestigatorProtective Becker & Morton yes Shin et al. yes Kitazawayes Epstein et al. yes Kass et al. yes Limitations of previous studies:  Varying endpoints  Limited treatment regimens  Small sample size

June, 2002 Ocular Hypertension Treatment Study (OHTS) Primary Goals  Evaluate the safety and efficacy of topical ocular hypotensive medication in delaying or preventing the development of POAG in individuals with elevated IOP  Identify baseline demographic and clinical factors that predict which participants will develop POAG

June, 2002 The OHTS Entry Criteria  Age  Normal visual fields –Humphrey 30-2  Normal optic discs  Untreated IOP: –24 to 32 mmHg in qualifying eye –21 to 32 mmHg in fellow eye

June, 2002 Patient found eligible for OHTS E ligible untreated IOPs on 2 visits 2 sets of normal & reliable HVFs per VFRC Optic discs judged normal by ODRC Medication Topical treatment to lower IOP 20% and IOP < 24 mm HgObservation No topical treatment to lower IOP Randomization Reproducible Abnormality 3 consecutive visual fields and/or 2 consecutive sets of optic disc photographs as determined by masked readers at ODRC or VFRC Monitoring Humphrey 30-2 q6 months Stereoscopic disc photos annually POAG Visual field and/or optic disc changes attributed to POAG by masked Endpoint Committee Adjust therapy if target not met

June, 2002 Baseline Characteristics by Randomization Group Gender & Age Medicationn=817Observationn=819 Male43.9%42.2% Female56.1%57.8% 40 to to %35.0% > 50 to %31.6% > 60 to %25.6% > 70 to %7.7% Ages

June, 2002 Medicationn=817Observation n=819 n=819 Native American 0.1%0.4% Asian0.5%1.2% African American 25.0%25.0% Hispanic2.9%4.3% Caucasian70.6%68.4% Other1.0%0.7% Baseline Characteristics by Randomization Group Self-designated Race

June, 2002 * Overall n=1398 for central corneal thickness, n=699 (86%) per randomization group. Measurements were conducted after 1999, about 2 years after the last participant was randomized. Baseline Characteristics by Randomization Group Ophthalmic Measurements Medicationn=817 (mean ± S.D.) Observation n=819 n=819 (mean ± S.D.) IOP (mm Hg) 24.9 ± ± 2.7 Cup:Disc Ratio (Horizontal) 0.36 ± ± 0.18 Cup:Disc Ratio (Vertical) 0.39 ± ± 0.19 Central Corneal Thickness (microns)* ± ± 37.7 Refraction (spherical equivalent in Diopters) ± ± 2.35

June, 2002 Baseline Characteristics by Randomization Group Visual Field Indices Medicationn=817 (mean ± S.D.) Observation n=819 n=819 (mean ± S.D.) Mean Deviation (dB) ± ± 1.03 Pattern Standard Deviation (dB) 1.92 ± ± 0.21 Corrected Pattern Standard Deviation (dB) 1.12 ± ± 0.36

June, 2002 Baseline Characteristics by Randomization Group Possible Risk Factors Medicationn=817Observation n=819 n=819 Prior use of Ocular Hypotensive Medication 35.0%39.3% First Degree Family History of Glaucoma 34.0%35.6% Myopia > 1 diopter Spherical Equivalent 34.4%33.7% Oral Beta Adrenergic Antagonist 5.4%4.6% Oral Calcium Channel Blocker 12.8%14.0%

June, 2002 Baseline Characteristics by Randomization Group Medical History Medicationn=817Observation n=819 n=819 Migraine10.4%11.7% Diabetes11.5%12.1% Hypertension37.5%38.1% Low Blood Pressure 4.8%4.0% Cardiovascular Disease 5.8%6.5% Stroke0.9%1.6%

June, 2002 Medication IOP (mm Hg) Months Observation Box Plot of IOP by Randomization Group Median IOP is joined by a line. Box: 25% and 75% Whiskers: 10% and 90%

June, 2002 IOP By Race MedicationObservation African American n=203Othern=614 n=205Othern=614 IOP at baseline 25.1 ± ± ± ± 2.7 IOP averaged over scheduled follow- up visits 19.3 ± ± ± ± 2.8 Percent reduction from baseline ± 10% ± 10% -4.7 ± 13% -3.8 ± 11%

June, 2002 Percent of Medication Patients on Different Medications Patients may be on more than one medication Months

June, 2002 Progress and Outcome of Study Participants MedicationObservationAll n%n%n% Randomized Died Inactive Non-adherence to randomization Reproducible VF or Optic Disc abnormality due to any cause Endpoints attributed to POAG Log rank p<0.001

June, 2002 Primary POAG Endpoints* Log Rank P-value <0.001, Hazard Ratio 0.40, 95% CI (0.27, 0.59) *through 8 Nov 2001 Proportion POAG Medication Observation Months

June, 2002 Medication Observation *through 8 Nov 2001 Proportion POAG 1 st Visual Field POAG Endpoint* Log Rank P-value=0.002, Hazard Ratio=0.45, 95% CI (0.26, 0.76) Months

June, 2002 Medication Observation *through 8 Nov 2001 Proportion POAG 1 st Optic Disc POAG Endpoint* Log Rank P-value<0.001, Hazard Ratio 0.36, 95% CI (0.23, 0.56) Months

June, 2002 First POAG Endpoint per Participant MedicationObservation N%N% Visual Field Optic Disc Concurrent Visual Field and Optic Disc Total

June, 2002 All cause reproducible abnormalities in visual fields and/or optic discs were significantly reduced in medication group. Hazard ratio 0.58, 95% CI ( ) P=

June, 2002 Treatment perhaps less protective in African Americans African Americans  12.7% POAG endpoints in observation group  6.9% POAG endpoints in medication group  Hazard Ratio 0.54  P value for interaction 0.26 Others  10.2% POAG endpoints in observation group  3.6% POAG endpoints in medication group  Hazard Ratio 0.34

June, 2002 No Significant Safety Difference Between Randomization Groups  Mortality  Hospitalizations  New Medical Conditions  Worsening of Pre-existing Conditions  SF – 36/any subscale  Patient Reported Ocular and Systemic Symptoms

June, 2002 Percent Reporting Changes in Iris, Lids or Lashes Prostaglandin analog > 6 months n = % Observation group n = 631 8% P <0.001

June, 2002 No difference between randomization groups in serious AEs for 9 of 11 organ systems.

June, 2002 Borderline Safety Differences Between Randomization Groups  Cataract surgery  Serious psychiatric adverse events  Serious genitourinary adverse events

June, 2002 Summary  Treatment produced about a 20% reduction in IOP.  Treatment reduced incidence of POAG in OHT participants by more than 50% at 5 years from 9.5% in the Observation Group to 4.4 % in the Medication Group.  Little evidence of safety concerns.

June, 2002 Vertical C/D Ratio per 0.1 increase Horizontal C/D Ratio per 0.1 increase PSD per 0.2 dB increase CCT per 40 microns decrease IOP per mm Hg Heart Disease Diabetes Mellitus Male gender African American origin Age Decade Significant Baseline Predictive Factors from Univariate Proportional Hazards Models 1.43 (1.19, 1.71) Hazard Ratio (95% CI) 1.59 (1.09, 2.32) 1.87 (1.31, 2.67) 0.40 (0.18, 0.92) 2.11 (1.23, 3.62) 1.11 (1.04, 1.18) 1.88 (1.55, 2.29) 1.36 (1.16, 1.60) 1.25 (1.14, 1.38) 1.32 (1.19, 1.46)

June, 2002 Myopia Mean Deviation CPSD per 0.3 dB Stroke Low Blood Pressure High Blood Pressure Migraine Oral Calcium Channel Blocker Oral Beta Adrenergic Antagonists Family History Glaucoma Non Significant Baseline Predictive Factors from Univariate Proportional Hazards Models Hazard Ratio (95% CI) 1.10 (0.7, 1.59) 0.70 (0.26, 1.89) 1.35 (0.83, 2.19) 1.01 (0.58, 1.76) 1.31 (0.92, 1.87) 1.49 ( ) 1.42 (0.35, 5.75) 1.16 (0.99, 1.35) 0.86 (0.73, 1.02) (0.91 (0.62, 1.32)

June, 2002 Significant Baseline Predictive Factors from Multivariate Proportional Hazard Models Age (decade) Diabetes Mellitus IOP (per mmHg) CCT (per 40 µM decrease) PSD (per 0.2 dB increase) Horizontal C/D Ratio (per 0.1 increase) Vertical C/D Ratio (per 0.1 increase) 1.22 (1.01, 1.49) 0.37 (0.15, 0.90) 1.10 (1.04, 1.17) 1.71 (1.40, 2.09) 1.27 (1.06, 1.52) 1.27 (1.14, 1.40) 1.32 (1.19, 1.47) Hazard Ratio (95% CI)

June, 2002  African Americans have a higher prevalence and incidence of POAG.  OHTS data suggests that this racial effect may be due to thinner central corneas and larger cup/disc ratios.

June, 2002 POAG Endpoints by Central Corneal Thickness and Baseline IOP (mmHg) in Observation Group* Baseline IOP (mmHg) Central Corneal Thickness (microns) * through 8 Nov 2001 < >23.75 to < >25.75 < 555 >555 to < 588 >588 17%9%2% 12%10%7% 36%13%6%

June, 2002 Vertical C/D Ratio Central Corneal Thickness (microns) * through 8 Nov 2001 < 0.30 >0.30 to <0.50 >0.50 < 555>555 to < 588>588 15%1%4% 26%16%4% 22%16%8% POAG Endpoints by Central Corneal Thickness and Baseline Vertical C/D Ratio in Observation Group*

June, year-old WF  IOP24 / 24  C/D ratio0.1 vertical  Corneal thickness600 μ  Risk of POAG~ 1% / 5 years

June, year-old WF  IOP24 / 24  C/D ratio0.3  Corneal thickness540 μ  Risk of POAG ~ 7% / 5 years

June, year-old WF  IOP28 / 28  C/D ratio0.1  Corneal thickness600 μ  Risk of POAG~ 2% / 5 years

June, year-old WF  IOP24 / 24  C/D ratio0.5  Corneal thickness490 μ  Risk of POAG~ 20% / 5 years

June, year-old BM  IOP25 / 25  C/D ratio0.6  Corneal thickness510 μ  Risk of POAG~ 35% / 5 years

June, 2002 Strengths 1.Large sample size 2.Careful follow-up 3.Masked assessment of endpoints 4.Attribution of endpoints to cause by masked committee 5.Inclusion of all commercially available drugs 6.Careful quality control and feedback to technicians and photographers 7.True-incidence cases

June, 2002 Weaknesses 1.Convenience sample rather than population based 2.Relatively small number of POAG endpoints 3.Healthy volunteers 4.Limited IOP range 5.Limited to patients with reliable visual fields 6.“Squeaky clean” participants at baseline 7.High thresholds for endpoints 8.Some risk factors under-represented

June, 2002 Summary  Not every patient with OHT should be treated  Offer treatment to OHT patient at moderate to high risk taking into consideration:  Consider measuring corneal thickness in all patients with OHT or glaucoma.  Age  Medical status  Life expectancy  Likely treatment benefit

June, 2002 Possible Misinterpretations of OHTS 1.Treat all patients with elevated IOP. 2.Risk of POAG is low in this population. 3.Glaucoma medications are harmless. 4.Risk factors for developing POAG are clearly delineated; influence of race, gender, hypertension, heart disease, family history, blood pressure, and diabetes are all clear. 5.20% lowering of IOP is the correct target for OHT. 6.Drug X is proven to prevent glaucoma in OHT.

June, 2002 OHTS Resource Centers Study Chairman’s Office & Coordinating Center Washington University St. Louis, MO Visual Field Reading Center University of California, Davis Sacramento, CA Optic Disc Reading Center Bascom Palmer Eye Institute University of Miami Miami, FL

June, 2002 OHTS Clinical Centers  Bascom Palmer Eye Institute  Eye Consultants of Atlanta  Eye Physicians and Surgeons  Cullen Eye Institute  Devers Eye Institute  Emory Eye Institute  Henry Ford Hospitals  Johns Hopkins University  Krieger Eye Institute  Howard University  University of Maryland  University of California, Los Angeles  Charles Drew University  Kellogg Eye Center  Kresge Eye Institute  Great Lakes Eye Institute  University of Louisville  Mayo Clinic  New York Eye & Ear Infirmary  Ohio State University  Ophthalmic Surgeons & Consultants  Pennsylvania College of Optometry  MCP/Hahnemann University  Scheie Eye Institute  University of California, Davis  University of California, San Diego  University of California, San Francisco  University Suburban Health Center  University of Ophthalmic Consultants  Washington Eye Physicians & Surgeons  Eye Associates of Washington, DC  Washington University, St. Louis